Fix a case where AXTreeSerializer didn't handle reparenting.

Fix a case where AXTreeSerializer didn't handle reparenting.

Additional unit tests that caught this case are forthcoming. This is a
simple and safe patch for merging.

BUG=452607
Committed: https://crrev.com/7a6d7e70755232d232916d78ca8ae94a5ca14dea
Cr-Commit-Position: refs/heads/master@{#318660}

[dmazzoni, 2015-02-27 18:10:43]

dmazzoni@chromium.org changed reviewers:
+ dtseng@chromium.org

[David Tseng, 2015-02-27 19:00:47]

https://codereview.chromium.org/966833002/diff/1/ui/accessibility/ax_tree_ser...
File ui/accessibility/ax_tree_serializer.h (right):

https://codereview.chromium.org/966833002/diff/1/ui/accessibility/ax_tree_ser...
ui/accessibility/ax_tree_serializer.h:302: // that contains the old and new
parents of the reparented node.
Mention something about what's happening with the lca and the relationship it
has with the return value of AnyDescendantReparented. Also, state the reason why
we're doing this repeatedly.

https://codereview.chromium.org/966833002/diff/1/ui/accessibility/ax_tree_ser...
ui/accessibility/ax_tree_serializer.h:320: for (size_t i = 0; i <
client_lca->children.size(); ++i) {
I actually think we should call DeleteClientSubtree directly here (even for the
merge) since it makes things a lot clearer.

https://codereview.chromium.org/966833002/diff/1/ui/accessibility/ax_tree_ser...
ui/accessibility/ax_tree_serializer.h:331: // different now.
Not sure I like this comment here. Maybe put it outside/above the loop and
rephrase accordingly.

[David Tseng, 2015-02-27 19:19:43]

https://codereview.chromium.org/966833002/diff/1/ui/accessibility/ax_tree_ser...
File ui/accessibility/ax_tree_serializer.h (right):

https://codereview.chromium.org/966833002/diff/1/ui/accessibility/ax_tree_ser...
ui/accessibility/ax_tree_serializer.h:322:
DeleteClientSubtree(client_lca->children[i]);
nit: not for this cl, but this really should be s/Subtree/Descendants as I
understand the usage of these terms.

[dmazzoni, 2015-02-27 22:08:56]

Feedback addressed, take a look...

[David Tseng, 2015-02-28 17:37:05]

lgtm

[dmazzoni, 2015-03-01 03:54:15]

The CQ bit was checked by dmazzoni@chromium.org

[commit-bot: I haz the power, 2015-03-01 03:54:43]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/966833002/20001

[commit-bot: I haz the power, 2015-03-01 03:58:59]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-03-01 03:59:00]

Try jobs failed on following builders:
  ios_dbg_simulator_ninja on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/ios_dbg_simulator...)

[dmazzoni, 2015-03-02 06:02:46]

The CQ bit was checked by dmazzoni@chromium.org

[commit-bot: I haz the power, 2015-03-02 06:03:34]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/966833002/20001

[commit-bot: I haz the power, 2015-03-02 07:39:15]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2015-03-02 07:40:00]

Patchset 2 (id:??) landed as
https://crrev.com/7a6d7e70755232d232916d78ca8ae94a5ca14dea
Cr-Commit-Position: refs/heads/master@{#318660}